The writeup says the planet has 0.9 times Jupiter's mass, and the article says it has a temperature of 3100 F. This probably means it is not a true gas giant (whoopsie), but it is still hot enough to melt silicon and iron, so there's still no solid surface. Imagine a planet of magma. Were it to cool off, it should become a very large rocky object.
As it is, there's nothing to land on, and it's too hot for a ship to survive. And even if it were cool enough to safely land on, the gravity would be too high for human habitation.
Hmm. That paints an interesting picture. A few relatively common substances (like aluminum oxide) should be solid at those temperatures. Depending on their buoyancy relative to other components (mostly silicates), you might end up with a solid crust (modulo enough convection churn to make the San Andreas fault look like a nice picnic spot). This of course assumes that enough oxygen was bound into oxides in the first place, instead of mostly being bound as water and being stripped early in the planet's life/blown away to the outer system during accretion.
As for dropping probes into that ocean, an aluminum oxide shell should work quite well. Tungsten carbide-coated graphite might work too, being even more temperature-resistant, as long as it doesn't alloy with the surface material. It would be neat to try, though I doubt we'll get a chance any time soon:).
How would you set up the electronics/instruments on the probe, though? Diamond semiconductors, carbon or tungsten wires, aluminum oxide optics? It's an interesting thought experiment.
This depends strongly on the distance you choose to measure the force at. At a distance of 1m, as opposed to 1e-15m, the original ordering may be correct.
And as long as we're being nit-picky, I'll point out that human-observable phenomena tend to be larger than 1e-15m:).
It's a gas giant. Colony ships wouldn't be useful even if it was at a livable temperature. Granted, overly large moons might be habitable for some gas giants, but we'd have to be able to find them first.
Wouldn't this be more likely to be a very large rocky object? That close to the star, I'd expect volatiles to be stripped.
...wait. Isn't dark matter supposed to still have mass and gravity? So any dark matter of sufficient size would cause microlensing, not just baryons, no?
Only if it's concentrated into very compact clumps.
If it's a halo of low-mass chargeless particles, clumps would be much larger (think "galaxy-sized or bigger"). Such dark matter would not be directly visible within our galaxy, though "dark galaxies" or larger blobs of dark matter could be seen by lensing of distant quasars. Searches for "dark galaxies" have been underway for a while (though the main purpose is to look for ordinary galaxies that are just dim enough not to be seen that far away, if I understand correctly).
Things like "dark energy" might not clump at all (I'd have to look that up to confirm it, though).
Isn't dark matter simply matter that doesn't emit light? If stars get formed by huge clouds of gas that eventually create so much heat and pressure that it starts a process of fusion, then its more than likely all this dark matter we are talking about is just that, dark matter, dirt, whatever you want to call it.
It turns out that the measured effects of dark matter mean that only a small fraction of it can be "normal" matter. Look up "baryonic" and "non-baryonic" dark matter on Google for more information on the subject.
The "normal" component could be anything from white dwarf stars to brown dwarf super-planets to micro black holes to dust and gas, or all of the above. However, that still leaves most of the mass as something else.
By balance, he's probably talking about the balance of humans to plants and other animals. There exists in our biosphere a concept known as "Carrying Capacity of Species in an area." Look it up in your biology textbook.
As was described in my other messages in this thread, the carrying capacity for humans is not a constant - it is a function of both our lifestyle and of the industrial/technological effort we put into extending it. This makes it a very arbitrary number, whose exact value is chosen by our decisions as a society.
We're clearcutting forests to make room for mini-malls and parking lots all the time. The trees have a right to life too, don't they? Do you think there are too many trees? They don't grow at the expense of others like Humans do.
They grow at the expense of other types of tree, or of the other organisms that would flourish should the region in question not be forest. And vice versa. This is your own carrying capacity argument.
As far as "rights" are concerned, there are no fundammental "rights" dictated by nature, as the concept of "rights" is an artificial construct. Arguments based on "rights" reflect society's choices based on cultural values, not anything imposed externally. The "rights" of other organisms - or how many tigers and gorillas we want to have around, to use your other example - influence our _choice_ of where to put the balance point. Using them as support of a naturally-imposed balance point is silly, for the reasons outlined in this paragraph.
I sincerely hope that we as a society choose to leave vast swaths of untouched wilderness on the planet as parks, but it will be by charity, not by necessity for our survival.
If you exeed the carrying capacity of an area by increasing the population of one type of organism (man), 2 thing will happen: We won't have enough food to go around, and hunger is ALREADY a problem in the world, and Disease will be more prone - it's a proven fact.
These statements prove that we've already vastly increased the carrying capacity of the planet. Look at any major city; in a society with medieval technology, most of the population would die from cholera or some other disease due to poor sanitation, and cities far from farmland would starve due to vastly reduced crop yields and lack of volume of transportation.
Food production and resistance to disease have been greatly increased, increasing the carrying capacity. Thus, your arguments don't seem to support your original point.
Additionally, I'd like to disagree with your statement that Nature isn't responsible for trying to balance it's self out.
You seem to have misunderstood my point. Nature is extremely good at balancing itself out. My point is that the location of the balance point is arbitrary, being determined by the conditions nature is responding to. Thus, nature's balancing act says nothing about where it "should" be balanced.
Doesn't it seem like good logic that if Humans just went away for a while, Things would eventually return to the way they were before we were around?
And if trees went away for a while, ferns would re-inherit the earth. So what?
You're making the assumption that the state of the world prior to human existance is preferable to the state of the world with humans on it. This is a value judgement, based on your own personal values. As described above, this kind of value judgement is arbitrary, and so does not support the asseration that the world "should" be that way.
Humans should learn that if they want to survive, they shouldn't live in an area where the threat is prevalent.
We have the ability to alter our environment to make it more favourable to us. Why should we *not* use it? Ever since the first fire was lit in a cave, it's been done.
To conclude, you seem to be making the implicit assumption that the way the world was before humans were present is the way the world "should" be. I think this is silly.
For what it's worth, email handling is not usually a CPU-limited activity. On small systems, hardware limits don't really enter into it -- a smallish site can handle a normal mail load nicely on a 486! -- and on larger systems, tends to be I/O-limited, by either the speed of the network interfaces or that of the disks. Since it isn't CPU-limited, increasing the CPU load involved a little bit, by adding filtering, won't have all that much impact on the throughput.
I strongly suspect that Bayesian filtering would turn mail processing into a CPU-bound activity. You're converting words into known tokens, looking up coefficients associated with each distinct token, and then manipulating them. If anything, it resembles compiling as a workload.
To prove the issue either way, of course, I'd have to get off my tail and actually build an efficient filter and test it. As an O(n log n) problem, it _might_ not be CPU bound, for low enough disk/network throughput.
Disease is an entirely different beast, however. The "lag time" that you refer to is quite significant. With the same number of births, things will get a lot more crowded if the average person lives to sixty, rather than to twenty.
Population would increase by a constant factor (the ratio of the old and new lifespans). If we find a cure for *all* disease, this might be a factor of 10 (accidents would become the primary cause of death).
This is perfectly manageable.
Population problems arise when the birth rate is consistently higher than the death rate. As mentioned previously, this has nothing to do with lifespan, and everything to do with social norms.
In summary, I see no overpopulation threat from longevity.
Your argument about plagues is a red herring, as these are almost all confined to relatively small subsets of the population (people within the affected area), and so have virtually no impact on the population as a whole. Even a "tens of millions of people die" plague is barely a blip. Furthermore, the spreading of plagues on a global scale is caused by technology (the availability of cheap global transport), and so is as artificial as the cures for them. Thus, arguing that nature "wants" to stabilize our population through illness is still silly.
In summary, I do not find your disease argument convincing.
In the natural world, populations are limited by resource availability, whether that resource is habitat or food. With technology, both limits can be pushed back for humanity by many orders of magnitude. How far we push them back is, as described previously, an arbitrary choice based on what consequences we want to accept, as opposed to something imposed by nature.
I may sound like a horrible person here, but I really think that as soon as we start screwing around with nature, we throw the balance out the window. The human population is already way too large as it is.
I'll bite.
You are making several questionable assumptions in your post.
First of all, you're assuming that there is a size that the human population "should" be. How did you derive this value, and what was it? As far as I can see, the human race can survive at just about any population level it pleases - there's just a sliding scale of consequences, which in turn depends very strongly on _how_ people choose to run their lives. So both the desired population and the effects of maintaining this population are pretty arbitrary decisions.
Second of all, you're assuming that there is a "balance" that must be maintained. Historically, the ecosphere has done a very good job of maintaining itself despite far greater changes than humanity has wrought. There is a continuum of possible balance points, each with their own consequences - where we want to place the balance point is a decision, not something dictated by nature.
Much like developing cures for disease, stopping hurricanes from hitting population centers is just another way to screw over any form of population control.
Hurricanes do not contribute substantially to population control.
Neither does disease, really. We'll always die of _something_. The lag time is pretty much irrelevant over the long term. The period of fertility for women is pretty much the same, so people could live to age 300+ without affecting the number of children they had over the 20-year window. The number of children per couple is a social issue, not one of longevity.
Otherwise the carbonic acid would react with the aluminum, and leave you with a nasty taste (I believe due to Aluminum Oxide? but its been a while since high school Chemistry).
Aluminum oxide is not soluble and almost certainly doesn't have any taste (it's even more stable than silica).
What you get after dissolving aluminum with an acid is hydrogen and an aluminum-based salt. This would be aluminum carbonate for carbonic acid, and aluminum phosphate for the phosphoric acid many drinks use as a flavouring agent.
I left a case of coke unused for about 6 months once. Tasted very odd after the lining broke down.
I think you're missing the point. It does not matter at all how much potential energy is stored in a few kms of altitude. But if this craft is good at gliding, which it should be, being light with a large wingspan, it would only drop a few kilometers overnight through unpowered gliding.
I find it doubtful that it would only drop a few kilometres after gliding for 12 hours.
Back of the envelope calculation supporting this:
From the Helios page cited by another poster, Helios weighs 1600 lbs, and has 14 moters rated to 1.5 kW each (2 hp). This gives a power consumption of 0.029 kW/kg, or about 1.25 MJ/kg over the course of 12 hours. Dropping 10 km gives you 0.1 MJ/kg. Power used in flight is far greater than gravitational potential energy for any practical drop, even counting the fact that the motors are not perfectly efficient at driving the craft.
In summary, you'd hit the ocean quite early trying to glide overnight.
The builders of the craft seem to agree, as the project page mentiones fuel cells being _required_ for operation through the night.
The need for power overnight isn't to keep it powered overnight (are you thinking payload?) so much as it is to keep the whole thing aloft. Their site or somewhere said the plane consumes about 30 kW. Obviously, you can't use the engines to produce electricity and thrust simultaneously.
You could do what amounts to this by turning off the engines and just gliding down at a shallow angle (spending gravitational potential energy to maintain airspeed). I'd be very surprised if the craft would only sink a few km after gliding unpowered for up to 12 hours, though.
The energy obtainable by dropping a few kilometers -- hardly a big deal for a wing 40km up -- would be just as much as could be stored in fuel cells
This turns out not to be the case.
Energy stored gravitationally is F*d: 10N/kg * 1e3m, or 10 kJ per kg per km.
Energy density for conventional batteries is at least 10 times this. Energy density for chemical fuels is several hundred times this. So, for a fuel cell power storage system representing a small fraction of the craft's mass, you get much more power storage capacity than you'd get from having the craft sink and rise again.
The main problem will be keeping the weight of the hydrogen tank down (if stored at high pressure), or the volume down enough to fit in the craft's airframe (if stored at low pressure).
An email message (or packet) should be authenticated at its source as coming from a valid, certifyable and traceable source.
The problem with this is twofold: First, you're going to have a very difficult time getting people to agree on trustworthy sources, and second, you get the same problem as we have with DNS - the people who hold the keys have far too much power.
And unless all servers on the planet agree on a set of athentication servers, you'll still be able to inject spam into the system from remote relays (c.f. the china problem right now).
I'm not convinced this approach is practical. It's great in principle; I just don't think any likely implementation would work very well.
What could be better for a professional Spammer than attending an Anti-Spam Conference? Learn all the techniques and issues you will have to encounter in the upcoming months.
How would this help them? People have known how the RBL, for instance, works for years, and yet it's still quite effective.
Likewise, filtering based on content still works despite being around for a while because spam mails... have to contain spam.
In summary, I don't see what they'd learn that would be of use to them.
It appears that the only solution to eliminating SPAM is to develop a completely new architecture for handling email which would simply not provide mechanisms for the broadcast of SPAM, and the hijacking of mail servers.
How about just properly configuring the existing mailservers?
The hijacking problem is mainly with mail servers misconfigured as open relays.
No switchover needed.
As was pointed out in the last round of spam-article comments, you can't eliminate the header-forging problem, as at some point you have to trust the server that's supplying you with mail. So a new scheme would not help with this.
In summary, I don't see how switching to a new scheme would help.
I recently read in Discover magazine, that some astrophysicists are openly questioning whether we have the mental prowess to actually understand many of the mysteries in the universe.
For analogy, they talked about Apes. While it is clear that an Ape has intelligence, we do not expect them to start solving differential calculus any time soon. Their intelligence can't even conceive that such a thing exists.
While this is an interesting idea, I'm not worrying about it for two reasons:
We've repeatedly demonstrated the ability to augment our own intelligence
Writing does this by increasing the amount of state information we can deal with for a given problem (I can't multiply 100-digit numbers in my head, but I can on paper). Calculating machines - from the abacus on up - do this by giving us "co-processors" to handle tasks that our brains are not suited for. If there's a good argument for this augmentation having a fundamental limit, I haven't heard it yet.
If we don't understand it, something we build might.
If we postulate that we can build an artificial intelligence, and postulate further that we can build artificial intelligences that are smarter than we are, either that intelligence or one of its descendants may be able to grasp whatever arbitrarily complex model truly represents reality, if it can be grasped at all. The same argument applies if we genetically engineer creatures smarter than non-modified humans.
In summary, I think either we or our creations will likely be smart enough to understand the universe, if anything can.
While your post was interesting, there are a few statements you make that seem to be based on incomplete information:
The quantum leap will happen when enough detailed data is gathered about actual events as they happen, which can then be extrapolated to the past.
Unfortunately, it is unlikely that any possible measurements will allow this. Firstly, even if you assume a closed system (the solar system not being substantially affected by things outside it), small uncertainties in knowledge of the system's state at the time of measurement grow very rapidly as state is extrapolated forwards or backwards in time for complex systems (like the solar system). While some parts of the system may be insensitive to error (we can predict with reasonable certainty where Jupiter was a hundred thousand years ago), other parts aren't, and uncertainties in even stable parts still stack up over time.
Secondly, even with perfectly accurate measurements, the solar system (or anything else smaller than the universe) is a closed system. You'd need not only measurements of the piece you're interested in, but of all parts of the forward-facing light cone of the past state you're interested in... and then have some way to subtract the contributions from everything in the past light-cone of the area you're sampling to get the forward light-cone. And then you repeat the process for this larger sample set. So, you end up making approximations about the contribution of external events, as these cannot be known with certainty without knowing the state of the entire universe.
In summary, detailed, accurate prediction into the distant past or future is impossible
so the real question is what they do then... it's a bit easy, really, to take your model and add a couple of new variables in there until they get it right. This doesn't really prove anything though, does it?
Even an accurate model proves nothing. A model is a description of a system used as an aid in making predictions about the behavior of the system. The real way the system works may bear no relation to the structure of the model, even if the predictions seem perfectly accurate.
In practice, however, a model that makes many accurate predictions and very few inaccurate ones stands a good chance of being a reasonable approximation to the way reality works.
What we're doing by refining these models is trying to get a better understanding of how reality works. If experimental evidence is at odds with the model's predictions, of *course* it will be changed. However, as the model was already based on experimental evidence to the greatest degree possible, it still stands a reasonable chance of being mostly correct. Thus, it is modified, instead of thrown away and replaced.
To cause a model to be thrown away, you don't just have to show that it mispredicts some cases - you have to provide a replacement that's better than the original.
In summary, as long as the current system formation models are the most accurate of the models offered, we'll refine them, and not replace them.
The moon creation simulation is the one that gets me. They seem still to be assuming that it's ONE impact that created the moon, and even give the analogy of a small car crashing into an SUV (follow links from moon story). I think it's much more chaotic than that, and is really a big highway pile-up, but where some cars could still run, and were driven away billions of years ago, some have degraded into other rocks and asteroids, and the big bit in the middle coalesced into the moon. Three-body collisions between very large objects are far, far less common than two-body collisions. Space is big; the chances of even two large bodies being in the same place at the same time is remote. Three is even less likely.
If you postulate that collisions are frequent enough for three-body collisions to occur, then the inescapable conclusion is that any products of three-body collisions would be utterly changed by the far more frequent two-body collisions, making the existance of three-body collisions moot.
In summary, a two-body scenario for creation of the moon is the most likely.
I think it's way too complex for a computer to simulate; every atom has a/dev/random (OK it's more like a predictable Windows TCP/IP stack, but there's some entropy in there), and that's the real problem. How do you simulate all of those?
By realizing what parameters have a significant contribution to the simulation, and which don't. We can model the orbit of the earth extremely accurately without having to know the state of every atom within it; its travel is primarily affected by only its total mass and the position of its center of mass. Anyone proposing a model of a system or writing up the results of a paper based on a new simulation will explain in great detail why they only need to consider the parameters they do, and what the resulting error ranges will be. In summary, solar system simulations can be trusted to be reasonably useful without tracking the state of every atom in the solar system.
The real excitement comes when currently forming galaxies can be studied over a long enough period - perhaps by simultaneously studying several galaxies in enough detail to come up with decent fluid/gas dynamics in space.
Unfortunately, except for very special cases (like looking at the black holes at the hearts of galaxies), the distances and time scales involved prevent us from getting more than one snapshot of a galaxy's behavior. Galaxies are tens to hundreds of thousands of light-years wide. As most parts of them move far slower than light, the time required for any substantial galaxy-scale phenomenon to occur - even a very fast one, by galactic standards - will be many millions of years. It is unlikely that we will have time to observe this.
Galactic formation also finished many billions of years ago. The forming galaxies we can still observe are far enough away to be impractical to study (billions of light-years, to look back billions of years; objects at that distance appear as points only).
In summary, both distance and time concerns make the observation of large-scale changes in galaxies impractical for the forseeable future.
Re:Possible for transparent x86 emulation on Linux
on
Bochs 2.0 Released
·
· Score: 2
Anyway, I don't get what cool possibilities for a PhD you are seeing there, since a dissertation is supposed to be new research, not reimplementation of existing technology.
I wasn't aware that anyone had implemented a really good cross-compiling optimizing emulator, which would have made such a thing a viable research topic. My mistake. Expressing it in terms of PhDs was an attempt at pointing out exactly how much work is involved in developing such a thing, to forestall "can we tweak bochs to do this?" comments.
Re:Possible for transparent x86 emulation on Linux
on
Bochs 2.0 Released
·
· Score: 2
One thing Linux on non-x86 platforms lacks is transparent X86 emulation, like on the Macintosh with its transparent 68K emulation, you click on a 68K app and it just works. I should be able to run a X86 ELF image on a non-X86 Linux box and have it just WORK! The Bochs approach is not the best way, since it's a virtual machine and emulates everything. A better way would be for X86 emulation only when needed, such as the application program code itself (syscalls continue to use the native library)
Dealing with endianness issues when wrapping all possible system calls would be so horrible it's not funny. Too many calls, too much mucking about to see what's *really* endian-sensitive under what conditions, and things like driver IOCTLs that you just plain don't know whether to wrap or not.
OTOH, emulating x86 is a horrid screaming nightmare. The 68k architecture is relatively clean, relatively simple. i686 is, well, *not*. A clean, easy to maintain implementation runs extremely slowly. An implementation based on JIT cross-compiling and re-optimization of code improves to merely "crawling so slowly you want to claw your eyes out", as you have to track *all* possible side effects of all instructions, in an architecture that was definitely not designed to make that easy. If you're a god and write an emulator that not only cross-compiles but that tracks all side effects, finds out which ones don't matter and discards them, speculatively unrolls and optimizes and maybe even skips loops with code that checks for premature exits and state changes (to roll back state to non-unrolled/skipped loops in case of mispredicts), and in all other ways just extracts the salient computations being performed while discarding all busy-waiting and non-computation cruft, then it'll just be "slow".
This would be a really cool series of PhD topics for about a dozen skilled CS grad students. After 10-15 years of work, this might be do-able, and the cross-compiling/optimization technology developed would have many other applications.
In the meantime, recompiling is probably the way to go.
In summary, good, real-time x86 emulation is a "pick one" scenario at the moment.
The Crusoe doesn't count, as they're mapping to hardware specifically designed to emulate x86 machines.
True, radio communications just aren't going to cut it. We can pick up radio-type signals from stars, but these are... well, not to put too fine a point on it, fucking stars.
I seem to recall reading that Earth outshines the sun in certain radio bands. Citation lost to the mists of time.
You could beamcast signals to another star easily enough, especially with a (very large) space-based dish. The problem is aperture size, not source power per se (you want the beam to have low divergence). While optical transmission doesn't require as large a dish for a given divergence, it does require far more energy to be detectable. You have to be bright enough to put a minimum of about 10 photons per $sample_period per $detector_area at the destination star system to be detected, and visible photons are many orders of magnitude more energetic. (I'm assuming we're doing detection by correlating many samples, instead of trying to dump enough energy to outshine the Sun in one pulse).
Broadcasting instead of beamcasting, we'd need vastly more power to be detectable at all.
Oh, I agree... But I'm not talking about the surface, nor am I just talking about the bombs that exist... I'm talking about all the bombs, all the nuclear wastes, and anything else that we don't need that is radioactivly decaying
What part of "we'd need 10 trillion metric tonnes to generate the required amount of energy" aren't we getting through to you?
All of the radioactive waste we're likely to produce over our lifetime as a civilization is less than the required amount.
I'm not talking replacing the core... just adding to it's nuclear material.
My point was that you'd have to add an amount of material comparable to that in the Earth's core to provide an adequate heat flux (well, maybe a quarter that due to smaller Martian surface area).
This isn't a stalled core that needs to be kick-started. This is a core that just produces way too small an amount of energy. Even the Earth's core is almost certainly sub-critical, regardless of the story-du-jour on Slashdot.
In summary, I think a far larger amount of effort would be needed than you are assuming.
Slashdot Mirror
The writeup says the planet has 0.9 times Jupiter's mass, and the article says it has a temperature of 3100 F. This probably means it is not a true gas giant (whoopsie), but it is still hot enough to melt silicon and iron, so there's still no solid surface. Imagine a planet of magma. Were it to cool off, it should become a very large rocky object.
:).
As it is, there's nothing to land on, and it's too hot for a ship to survive. And even if it were cool enough to safely land on, the gravity would be too high for human habitation.
Hmm. That paints an interesting picture. A few relatively common substances (like aluminum oxide) should be solid at those temperatures. Depending on their buoyancy relative to other components (mostly silicates), you might end up with a solid crust (modulo enough convection churn to make the San Andreas fault look like a nice picnic spot). This of course assumes that enough oxygen was bound into oxides in the first place, instead of mostly being bound as water and being stripped early in the planet's life/blown away to the outer system during accretion.
As for dropping probes into that ocean, an aluminum oxide shell should work quite well. Tungsten carbide-coated graphite might work too, being even more temperature-resistant, as long as it doesn't alloy with the surface material. It would be neat to try, though I doubt we'll get a chance any time soon
How would you set up the electronics/instruments on the probe, though? Diamond semiconductors, carbon or tungsten wires, aluminum oxide optics? It's an interesting thought experiment.
Actually, you got the order wrong.
:).
Strong, Electromagnetic, Weak, Gravitational
This depends strongly on the distance you choose to measure the force at. At a distance of 1m, as opposed to 1e-15m, the original ordering may be correct.
And as long as we're being nit-picky, I'll point out that human-observable phenomena tend to be larger than 1e-15m
It's a gas giant. Colony ships wouldn't be useful even if it was at a livable temperature. Granted, overly large moons might be habitable for some gas giants, but we'd have to be able to find them first.
Wouldn't this be more likely to be a very large rocky object? That close to the star, I'd expect volatiles to be stripped.
...wait. Isn't dark matter supposed to still have mass and gravity? So any dark matter of sufficient size would cause microlensing, not just baryons, no?
Only if it's concentrated into very compact clumps.
If it's a halo of low-mass chargeless particles, clumps would be much larger (think "galaxy-sized or bigger"). Such dark matter would not be directly visible within our galaxy, though "dark galaxies" or larger blobs of dark matter could be seen by lensing of distant quasars. Searches for "dark galaxies" have been underway for a while (though the main purpose is to look for ordinary galaxies that are just dim enough not to be seen that far away, if I understand correctly).
Things like "dark energy" might not clump at all (I'd have to look that up to confirm it, though).
Isn't dark matter simply matter that doesn't emit light? If stars get formed by huge clouds of gas that eventually create so much heat and pressure that it starts a process of fusion, then its more than likely all this dark matter we are talking about is just that, dark matter, dirt, whatever you want to call it.
It turns out that the measured effects of dark matter mean that only a small fraction of it can be "normal" matter. Look up "baryonic" and "non-baryonic" dark matter on Google for more information on the subject.
The "normal" component could be anything from white dwarf stars to brown dwarf super-planets to micro black holes to dust and gas, or all of the above. However, that still leaves most of the mass as something else.
By balance, he's probably talking about the balance of humans to plants and other animals. There exists in our biosphere a concept known as "Carrying Capacity of Species in an area." Look it up in your biology textbook.
As was described in my other messages in this thread, the carrying capacity for humans is not a constant - it is a function of both our lifestyle and of the industrial/technological effort we put into extending it. This makes it a very arbitrary number, whose exact value is chosen by our decisions as a society.
We're clearcutting forests to make room for mini-malls and parking lots all the time. The trees have a right to life too, don't they? Do you think there are too many trees? They don't grow at the expense of others like Humans do.
They grow at the expense of other types of tree, or of the other organisms that would flourish should the region in question not be forest. And vice versa. This is your own carrying capacity argument.
As far as "rights" are concerned, there are no fundammental "rights" dictated by nature, as the concept of "rights" is an artificial construct. Arguments based on "rights" reflect society's choices based on cultural values, not anything imposed externally. The "rights" of other organisms - or how many tigers and gorillas we want to have around, to use your other example - influence our _choice_ of where to put the balance point. Using them as support of a naturally-imposed balance point is silly, for the reasons outlined in this paragraph.
I sincerely hope that we as a society choose to leave vast swaths of untouched wilderness on the planet as parks, but it will be by charity, not by necessity for our survival.
If you exeed the carrying capacity of an area by increasing the population of one type of organism (man), 2 thing will happen: We won't have enough food to go around, and hunger is ALREADY a problem in the world, and Disease will be more prone - it's a proven fact.
These statements prove that we've already vastly increased the carrying capacity of the planet. Look at any major city; in a society with medieval technology, most of the population would die from cholera or some other disease due to poor sanitation, and cities far from farmland would starve due to vastly reduced crop yields and lack of volume of transportation.
Food production and resistance to disease have been greatly increased, increasing the carrying capacity. Thus, your arguments don't seem to support your original point.
Additionally, I'd like to disagree with your statement that Nature isn't responsible for trying to balance it's self out.
You seem to have misunderstood my point. Nature is extremely good at balancing itself out. My point is that the location of the balance point is arbitrary, being determined by the conditions nature is responding to. Thus, nature's balancing act says nothing about where it "should" be balanced.
Doesn't it seem like good logic that if Humans just went away for a while, Things would eventually return to the way they were before we were around?
And if trees went away for a while, ferns would re-inherit the earth. So what?
You're making the assumption that the state of the world prior to human existance is preferable to the state of the world with humans on it. This is a value judgement, based on your own personal values. As described above, this kind of value judgement is arbitrary, and so does not support the asseration that the world "should" be that way.
Humans should learn that if they want to survive, they shouldn't live in an area where the threat is prevalent.
We have the ability to alter our environment to make it more favourable to us. Why should we *not* use it? Ever since the first fire was lit in a cave, it's been done.
To conclude, you seem to be making the implicit assumption that the way the world was before humans were present is the way the world "should" be. I think this is silly.
For what it's worth, email handling is not usually a CPU-limited activity. On small systems, hardware limits don't really enter into it -- a smallish site can handle a normal mail load nicely on a 486! -- and on larger systems, tends to be I/O-limited, by either the speed of the network interfaces or that of the disks. Since it isn't CPU-limited, increasing the CPU load involved a little bit, by adding filtering, won't have all that much impact on the throughput.
I strongly suspect that Bayesian filtering would turn mail processing into a CPU-bound activity. You're converting words into known tokens, looking up coefficients associated with each distinct token, and then manipulating them. If anything, it resembles compiling as a workload.
To prove the issue either way, of course, I'd have to get off my tail and actually build an efficient filter and test it. As an O(n log n) problem, it _might_ not be CPU bound, for low enough disk/network throughput.
I left a case of coke unused for about 6 months once. Tasted very odd after the lining broke down.
Did you die from drinking it?
Not yet, though I suppose the jury's still out.
Strangely, it gave one heck of a caffeine/sugar rush. Not that I'm about to repeat the experiment.
Disease is an entirely different beast, however. The "lag time" that you refer to is quite significant. With the same number of births, things will get a lot more crowded if the average person lives to sixty, rather than to twenty.
Population would increase by a constant factor (the ratio of the old and new lifespans). If we find a cure for *all* disease, this might be a factor of 10 (accidents would become the primary cause of death).
This is perfectly manageable.
Population problems arise when the birth rate is consistently higher than the death rate. As mentioned previously, this has nothing to do with lifespan, and everything to do with social norms.
In summary, I see no overpopulation threat from longevity.
Your argument about plagues is a red herring, as these are almost all confined to relatively small subsets of the population (people within the affected area), and so have virtually no impact on the population as a whole. Even a "tens of millions of people die" plague is barely a blip. Furthermore, the spreading of plagues on a global scale is caused by technology (the availability of cheap global transport), and so is as artificial as the cures for them. Thus, arguing that nature "wants" to stabilize our population through illness is still silly.
In summary, I do not find your disease argument convincing.
In the natural world, populations are limited by resource availability, whether that resource is habitat or food. With technology, both limits can be pushed back for humanity by many orders of magnitude. How far we push them back is, as described previously, an arbitrary choice based on what consequences we want to accept, as opposed to something imposed by nature.
No extra hardware at all
You do realize that calculating spam-likelihood probabilities requires nonzero amounts of processing power/cpu time, right?
I may sound like a horrible person here, but I really think that as soon as we start screwing around with nature, we throw the balance out the window. The human population is already way too large as it is.
I'll bite.
You are making several questionable assumptions in your post.
First of all, you're assuming that there is a size that the human population "should" be. How did you derive this value, and what was it? As far as I can see, the human race can survive at just about any population level it pleases - there's just a sliding scale of consequences, which in turn depends very strongly on _how_ people choose to run their lives. So both the desired population and the effects of maintaining this population are pretty arbitrary decisions.
Second of all, you're assuming that there is a "balance" that must be maintained. Historically, the ecosphere has done a very good job of maintaining itself despite far greater changes than humanity has wrought. There is a continuum of possible balance points, each with their own consequences - where we want to place the balance point is a decision, not something dictated by nature.
Much like developing cures for disease, stopping hurricanes from hitting population centers is just another way to screw over any form of population control.
Hurricanes do not contribute substantially to population control.
Neither does disease, really. We'll always die of _something_. The lag time is pretty much irrelevant over the long term. The period of fertility for women is pretty much the same, so people could live to age 300+ without affecting the number of children they had over the 20-year window. The number of children per couple is a social issue, not one of longevity.
In summary, your argument makes no sense.
Otherwise the carbonic acid would react with the aluminum, and leave you with a nasty taste (I believe due to Aluminum Oxide? but its been a while since high school Chemistry).
Aluminum oxide is not soluble and almost certainly doesn't have any taste (it's even more stable than silica).
What you get after dissolving aluminum with an acid is hydrogen and an aluminum-based salt. This would be aluminum carbonate for carbonic acid, and aluminum phosphate for the phosphoric acid many drinks use as a flavouring agent.
I left a case of coke unused for about 6 months once. Tasted very odd after the lining broke down.
I think you're missing the point. It does not matter at all how much potential energy is stored in a few kms of altitude. But if this craft is good at gliding, which it should be, being light with a large wingspan, it would only drop a few kilometers overnight through unpowered gliding.
I find it doubtful that it would only drop a few kilometres after gliding for 12 hours.
Back of the envelope calculation supporting this:
From the Helios page cited by another poster, Helios weighs 1600 lbs, and has 14 moters rated to 1.5 kW each (2 hp). This gives a power consumption of 0.029 kW/kg, or about 1.25 MJ/kg over the course of 12 hours. Dropping 10 km gives you 0.1 MJ/kg. Power used in flight is far greater than gravitational potential energy for any practical drop, even counting the fact that the motors are not perfectly efficient at driving the craft.
In summary, you'd hit the ocean quite early trying to glide overnight.
The builders of the craft seem to agree, as the project page mentiones fuel cells being _required_ for operation through the night.
The need for power overnight isn't to keep it powered overnight (are you thinking payload?) so much as it is to keep the whole thing aloft. Their site or somewhere said the plane consumes about 30 kW. Obviously, you can't use the engines to produce electricity and thrust simultaneously.
You could do what amounts to this by turning off the engines and just gliding down at a shallow angle (spending gravitational potential energy to maintain airspeed). I'd be very surprised if the craft would only sink a few km after gliding unpowered for up to 12 hours, though.
The energy obtainable by dropping a few kilometers -- hardly a big deal for a wing 40km up -- would be just as much as could be stored in fuel cells
This turns out not to be the case.
Energy stored gravitationally is F*d: 10N/kg * 1e3m, or 10 kJ per kg per km.
Energy density for conventional batteries is at least 10 times this. Energy density for chemical fuels is several hundred times this. So, for a fuel cell power storage system representing a small fraction of the craft's mass, you get much more power storage capacity than you'd get from having the craft sink and rise again.
The main problem will be keeping the weight of the hydrogen tank down (if stored at high pressure), or the volume down enough to fit in the craft's airframe (if stored at low pressure).
An email message (or packet) should be authenticated at its source as coming from a valid, certifyable and traceable source.
The problem with this is twofold: First, you're going to have a very difficult time getting people to agree on trustworthy sources, and second, you get the same problem as we have with DNS - the people who hold the keys have far too much power.
And unless all servers on the planet agree on a set of athentication servers, you'll still be able to inject spam into the system from remote relays (c.f. the china problem right now).
I'm not convinced this approach is practical. It's great in principle; I just don't think any likely implementation would work very well.
What could be better for a professional Spammer than attending an Anti-Spam Conference? Learn all the techniques and issues you will have to encounter in the upcoming months.
... have to contain spam.
How would this help them? People have known how the RBL, for instance, works for years, and yet it's still quite effective.
Likewise, filtering based on content still works despite being around for a while because spam mails
In summary, I don't see what they'd learn that would be of use to them.
It appears that the only solution to eliminating SPAM is to develop a completely new architecture for handling email which would simply not provide mechanisms for the broadcast of SPAM, and the hijacking of mail servers.
How about just properly configuring the existing mailservers?
The hijacking problem is mainly with mail servers misconfigured as open relays.
No switchover needed.
As was pointed out in the last round of spam-article comments, you can't eliminate the header-forging problem, as at some point you have to trust the server that's supplying you with mail. So a new scheme would not help with this.
In summary, I don't see how switching to a new scheme would help.
For analogy, they talked about Apes. While it is clear that an Ape has intelligence, we do not expect them to start solving differential calculus any time soon. Their intelligence can't even conceive that such a thing exists.
While this is an interesting idea, I'm not worrying about it for two reasons:
Writing does this by increasing the amount of state information we can deal with for a given problem (I can't multiply 100-digit numbers in my head, but I can on paper). Calculating machines - from the abacus on up - do this by giving us "co-processors" to handle tasks that our brains are not suited for. If there's a good argument for this augmentation having a fundamental limit, I haven't heard it yet.
If we postulate that we can build an artificial intelligence, and postulate further that we can build artificial intelligences that are smarter than we are, either that intelligence or one of its descendants may be able to grasp whatever arbitrarily complex model truly represents reality, if it can be grasped at all.
The same argument applies if we genetically engineer creatures smarter than non-modified humans.
In summary, I think either we or our creations will likely be smart enough to understand the universe, if anything can.
While your post was interesting, there are a few statements you make that seem to be based on incomplete information:
/dev/random (OK it's more like a predictable Windows TCP/IP stack, but there's some entropy in there), and that's the real problem. How do you simulate all of those?
The quantum leap will happen when enough detailed data is gathered about actual events as they happen, which can then be extrapolated to the past.
Unfortunately, it is unlikely that any possible measurements will allow this. Firstly, even if you assume a closed system (the solar system not being substantially affected by things outside it), small uncertainties in knowledge of the system's state at the time of measurement grow very rapidly as state is extrapolated forwards or backwards in time for complex systems (like the solar system). While some parts of the system may be insensitive to error (we can predict with reasonable certainty where Jupiter was a hundred thousand years ago), other parts aren't, and uncertainties in even stable parts still stack up over time.
Secondly, even with perfectly accurate measurements, the solar system (or anything else smaller than the universe) is a closed system. You'd need not only measurements of the piece you're interested in, but of all parts of the forward-facing light cone of the past state you're interested in... and then have some way to subtract the contributions from everything in the past light-cone of the area you're sampling to get the forward light-cone. And then you repeat the process for this larger sample set. So, you end up making approximations about the contribution of external events, as these cannot be known with certainty without knowing the state of the entire universe.
In summary, detailed, accurate prediction into the distant past or future is impossible
so the real question is what they do then... it's a bit easy, really, to take your model and add a couple of new variables in there until they get it right. This doesn't really prove anything though, does it?
Even an accurate model proves nothing. A model is a description of a system used as an aid in making predictions about the behavior of the system. The real way the system works may bear no relation to the structure of the model, even if the predictions seem perfectly accurate.
In practice, however, a model that makes many accurate predictions and very few inaccurate ones stands a good chance of being a reasonable approximation to the way reality works.
What we're doing by refining these models is trying to get a better understanding of how reality works. If experimental evidence is at odds with the model's predictions, of *course* it will be changed. However, as the model was already based on experimental evidence to the greatest degree possible, it still stands a reasonable chance of being mostly correct. Thus, it is modified, instead of thrown away and replaced.
To cause a model to be thrown away, you don't just have to show that it mispredicts some cases - you have to provide a replacement that's better than the original.
In summary, as long as the current system formation models are the most accurate of the models offered, we'll refine them, and not replace them.
The moon creation simulation is the one that gets me. They seem still to be assuming that it's ONE impact that created the moon, and even give the analogy of a small car crashing into an SUV (follow links from moon story). I think it's much more chaotic than that, and is really a big highway pile-up, but where some cars could still run, and were driven away billions of years ago, some have degraded into other rocks and asteroids, and the big bit in the middle coalesced into the moon.
Three-body collisions between very large objects are far, far less common than two-body collisions. Space is big; the chances of even two large bodies being in the same place at the same time is remote. Three is even less likely.
If you postulate that collisions are frequent enough for three-body collisions to occur, then the inescapable conclusion is that any products of three-body collisions would be utterly changed by the far more frequent two-body collisions, making the existance of three-body collisions moot.
In summary, a two-body scenario for creation of the moon is the most likely.
I think it's way too complex for a computer to simulate; every atom has a
By realizing what parameters have a significant contribution to the simulation, and which don't. We can model the orbit of the earth extremely accurately without having to know the state of every atom within it; its travel is primarily affected by only its total mass and the position of its center of mass. Anyone proposing a model of a system or writing up the results of a paper based on a new simulation will explain in great detail why they only need to consider the parameters they do, and what the resulting error ranges will be.
In summary, solar system simulations can be trusted to be reasonably useful without tracking the state of every atom in the solar system.
The real excitement comes when currently forming galaxies can be studied over a long enough period - perhaps by simultaneously studying several galaxies in enough detail to come up with decent fluid/gas dynamics in space.
Unfortunately, except for very special cases (like looking at the black holes at the hearts of galaxies), the distances and time scales involved prevent us from getting more than one snapshot of a galaxy's behavior. Galaxies are tens to hundreds of thousands of light-years wide. As most parts of them move far slower than light, the time required for any substantial galaxy-scale phenomenon to occur - even a very fast one, by galactic standards - will be many millions of years. It is unlikely that we will have time to observe this.
Galactic formation also finished many billions of years ago. The forming galaxies we can still observe are far enough away to be impractical to study (billions of light-years, to look back billions of years; objects at that distance appear as points only).
In summary, both distance and time concerns make the observation of large-scale changes in galaxies impractical for the forseeable future.
Anyway, I don't get what cool possibilities for a PhD you are seeing there, since a dissertation is supposed to be new research, not reimplementation of existing technology.
I wasn't aware that anyone had implemented a really good cross-compiling optimizing emulator, which would have made such a thing a viable research topic. My mistake. Expressing it in terms of PhDs was an attempt at pointing out exactly how much work is involved in developing such a thing, to forestall "can we tweak bochs to do this?" comments.
One thing Linux on non-x86 platforms lacks is transparent X86 emulation, like on the Macintosh with its transparent 68K emulation, you click on a 68K app and it just works. I should be able to run a X86 ELF image on a non-X86 Linux box and have it just WORK! The Bochs approach is not the best way, since it's a virtual machine and emulates everything. A better way would be for X86 emulation only when needed, such as the application program code itself (syscalls continue to use the native library)
Dealing with endianness issues when wrapping all possible system calls would be so horrible it's not funny. Too many calls, too much mucking about to see what's *really* endian-sensitive under what conditions, and things like driver IOCTLs that you just plain don't know whether to wrap or not.
OTOH, emulating x86 is a horrid screaming nightmare. The 68k architecture is relatively clean, relatively simple. i686 is, well, *not*. A clean, easy to maintain implementation runs extremely slowly. An implementation based on JIT cross-compiling and re-optimization of code improves to merely "crawling so slowly you want to claw your eyes out", as you have to track *all* possible side effects of all instructions, in an architecture that was definitely not designed to make that easy. If you're a god and write an emulator that not only cross-compiles but that tracks all side effects, finds out which ones don't matter and discards them, speculatively unrolls and optimizes and maybe even skips loops with code that checks for premature exits and state changes (to roll back state to non-unrolled/skipped loops in case of mispredicts), and in all other ways just extracts the salient computations being performed while discarding all busy-waiting and non-computation cruft, then it'll just be "slow".
This would be a really cool series of PhD topics for about a dozen skilled CS grad students. After 10-15 years of work, this might be do-able, and the cross-compiling/optimization technology developed would have many other applications.
In the meantime, recompiling is probably the way to go.
In summary, good, real-time x86 emulation is a "pick one" scenario at the moment.
The Crusoe doesn't count, as they're mapping to hardware specifically designed to emulate x86 machines.
True, radio communications just aren't going to cut it. We can pick up radio-type signals from stars, but these are... well, not to put too fine a point on it, fucking stars.
I seem to recall reading that Earth outshines the sun in certain radio bands. Citation lost to the mists of time.
You could beamcast signals to another star easily enough, especially with a (very large) space-based dish. The problem is aperture size, not source power per se (you want the beam to have low divergence). While optical transmission doesn't require as large a dish for a given divergence, it does require far more energy to be detectable. You have to be bright enough to put a minimum of about 10 photons per $sample_period per $detector_area at the destination star system to be detected, and visible photons are many orders of magnitude more energetic. (I'm assuming we're doing detection by correlating many samples, instead of trying to dump enough energy to outshine the Sun in one pulse).
Broadcasting instead of beamcasting, we'd need vastly more power to be detectable at all.
Oh, I agree... But I'm not talking about the surface, nor am I just talking about the bombs that exist... I'm talking about all the bombs, all the nuclear wastes, and anything else that we don't need that is radioactivly decaying
What part of "we'd need 10 trillion metric tonnes to generate the required amount of energy" aren't we getting through to you?
All of the radioactive waste we're likely to produce over our lifetime as a civilization is less than the required amount.
I'm not talking replacing the core... just adding to it's nuclear material.
My point was that you'd have to add an amount of material comparable to that in the Earth's core to provide an adequate heat flux (well, maybe a quarter that due to smaller Martian surface area).
This isn't a stalled core that needs to be kick-started. This is a core that just produces way too small an amount of energy. Even the Earth's core is almost certainly sub-critical, regardless of the story-du-jour on Slashdot.
In summary, I think a far larger amount of effort would be needed than you are assuming.